Nadav Ravid

Sound reaches people faster than anything else on a screen, and it usually gets the least thought.

The work starts as research rather than composition. Where will this actually be heard? What is the user doing at that moment, and what do they need to know? What is the brand claiming about itself? Only after that does it make sense to ask what a single button should sound like.

With both Gett and Pango, the useful outcome wasn't a full sonic identity. It was cracking the one sound the app most needed, and deriving it from brand values, user need and market research.

Gett, 2025. I rebuilt the sounds in the driver app. The whole family came out of one question: what does an accepted ride sound like from the driver's side? Entering a taxi is like entering the driver's living room, so a passenger asking to come in should sound like a knock at the door.

Pango, 2026. The same method produced a different answer for Israel's parking and mobility app. Activating parking or scanning onto a bus is the start of going somewhere, so the sound borrows from open-world games, where finding something pushes you onward instead of closing the loop.

Alongside project work I advise companies on audio and music strategy. I currently work with SNIP, which connects music creators with vetted industry experts for feedback on their tracks.

I got my first synthesiser at 13 and played my first party at 22. Neither has worn off - making music, or playing music I love to a room of people who are enjoying it.

I've been releasing under various names since 1997. I run Witness Protection Program, a small label that began in 2021 as an experiment in releasing anonymously, and before that I co-ran the label Botanika. Since 2021 I've also co-run a party in Tel Aviv where the music never goes above 120 BPM.

I lecture on music, culture and technology, mostly in Hebrew.

One talk works in English: Did You Pack This Yourself? It looks at where the music industry actually is with AI tools, why a lot of younger makers won't touch them, and how to use them without losing your own voice.

I also write a newsletter, Nadav Love Read, in Hebrew, about most of the above.

Read it on Substack ↗

I've been on both sides of this argument.

I built music discovery at iMesh and BearShare, and in 2014 consulted for Microsoft on its music recommendation research - work that rests on the premise that given enough signal, a system can tell you what you'll want to hear next. In 2011, between those two, I co-founded KZradio, an independent internet station, on the opposite premise: people with taste, choosing in public.

In 2015 I became the general manager of GLGLZ. The call I made there was that a national station shouldn't try to out-recommend a recommendation engine, because it will lose that fight. What it has instead is human judgement, and the job is to put that judgement in front rather than behind. We relaunched the schedule around presenters as personas in January 2016, two years before Spotify opened in Israel.

What holds my attention now is which parts of making something were ever really about efficiency, and which parts were the point.

I'm based in Israel and work with clients wherever they are. Email is the fastest way to reach me.